BAU projections for Europe with Eastern and Southern neighbouring countries

An approach with a Business-as-Usual scenario projection to 2020 for the Covenant of Mayors

from the Eastern Partnership

Greet Janssens-Maenhout, Ana Meijide-Orive, Diego Guizzardi, Valerio Pagliari, Andreea Iancu

EUR 25315 EN - 2012

BAU: 2020 ton CO2 /100km2

2

European Commission Joint Research Centre Institute for Environment and Sustainability Contact information Greet Janssens-Maenhout Address: Via E. Fermi, 2749 E-mail: greet.maenhout@jrc.ec.europa.eu Tel: +390332785831 Fax: +390332785837 http://ies.jrc.ec.europa.eu/ http://www.jrc.ec.europa.eu/ Legal Notice Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use which might be made of this publication.

Europe Direct is a service to help you find answers to your questions about the European Union

Freephone number (*):

00 800 6 7 8 9 10 11

(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.

A great deal of additional information on the European Union is available on the Internet. It can be accessed through the Europa server http://europa.eu/ JRC70917 EUR 25315 ISSN 1831-9424 ISBN 978-92-79-24803-0 doi:10.2788/26047 Luxembourg: Publications Office of the European Union © European Union, 2012 Reproduction is authorised provided the source is acknowledged Printed in Italy

3

COVENANT OF MAYORS EAST: AN APPROACH WITH A BUSINESS-AS-USUAL SCENARIO

PROJECTION TO 2020 FOR THE EASTERN PARTNERSHIP

G. Janssens-Maenhout, A. Meijide-Orive, D. Guizzardi, V. Pagliari, A. Iancu Joint Research Centre Ispra

Institute for Environment and Sustainability

May 2012

In response to the request of DG DEVCO for the Administrative Arrangement with JRC-IET on the Covenant of Mayors East

ABSTRACT

The methodology for the Covenant of Mayors – East needed to be extended with a business-as-usual projection of the emissions for 2020, from which national coefficients for the previous years are derived. In this way, signatories will be able to do their emission inventories of the present situation, and estimate which their emissions in 2020 will be. Then they will commit to an emission reduction target based on their projections of emissions for 2020 following the business-as-usual scenario. The factors are country-specific, calculated both for CO2 and CO2eq (CO2, CH4, N2O using the GWP100metric) in order to allow signatories to choose the approach they prefer. Moreover an urban dimension is provided, providing a margin on the projections.

4

CONTENT

1. Introduction 1.1. Geocoverage of the Covenant of Mayors East Programme 1.2. Transferring the Origin of the CoM EU Initiative 1.3. Proposed Adaption of the Methodology for Central Asian Cities 1.4. Local Projections: An Alternative?

2. Description of the Business-As-Usual (BAU) Scenario

2.1. Data and Models Used 2.2. Outcome of the Emission Projections

3. National coefficients from a BAU Projection for the CoM East Countries

3.1. Results for CO2 and CO2 eq national projections 3.2. Discussion of the limitations of the projections

4. The Urban Dimension 4.1. Distinction rural versus urban population 4.2. The resulting urban factor for the Eastern Partnership countries 4.3. Discussion and limitations of the use of the urban factor

5. Conclusion

References

Annex I: Ranking of world countries by GHGs Annex II: Ranking of world countries by CO2/cap Annex III: Ranking of world countries by CO2eq/cap List of Abbreviations

5

1. Introduction 1.1 Geocoverage of the Covenant of Mayors East Programme Within the Covenant of Mayors East programme, aspects need to be revised in the methodology of the original Covenant of Mayors EU, as described in the Guidebook "How to develop a Sustainable Energy Action Plan (SEAP)". This revision aims to tackle in a more appropriate way the specific institutional and economic situation of the 11 post-Soviet countries involved in the initiative, which in particular face lack of resources, absence of national framework, and an aspiration to strong economic growth (recuperating from the recession after the breakdown of the Soviet Union). The countries included in the CoM – East are the following:

• Armenia (ARM) • Azerbaijan (AZE) • Belarus (BLR) • Georgia (GEO) • Kazakhstan (KAZ) • Kyrgyzstan (KGZ) • Moldova (MDA) • Tajikistan (TJK) • Turkmenistan (TKM) • Ukraine (UKR) • Uzbekistan (UZB)

It should be considered that the situation between these countries is quite different from that of the EU. For these 11 post-soviet countries this reduction should be specified based on a baseline year emission inventory and the signatories will be given the possibility to use a business-as-usual (BAU) scenario to estimate their emissions in 2020. They will be allowed to use their own way to estimate their emissions in 2020 or to use factors provided by JRC. This second option would avoid a burden to the signatories in their economic aspiration. National coefficients are derived the European Commission’s Joint Research Centre, based on the energy consumption projections with an in-house EC model for energy-related activity increase. As for the baseline emission inventory (BEI) is recommended a recent year (later than 1990), which is representative for the current economic situation and for which reliable statistical data are available. The BEI should include the key sectors (residential and transport) as defined for the Covenant of Mayors initiative.

1.2 Transferring the origin of the Covenant of Mayors EU initiative

Key of the Covenant of Mayors methodology is the Sustainable Energy Action Plan (SEAP), in which signatories commit to a minimum CO2 emission reduction target of 20% by 2020 and define the actions they need to put in place to reach their commitment. A more specific overview of the original initiative as defined for the Covenant of Mayors 2020 can be found in Covenant of Mayors core text and is briefly described underneath. A city who signs up the Covenant of Mayors commits to:

reduce the CO2 emissions in its territory by at least 20% by 2020. prepare a Baseline Emission Inventory (BEI) as a basis for the SEAP. submit the SEAP, officially approved by the Local Authority, within the year following the

adhesion to the Covenant of Mayors. To elaborate and implement a successful SEAP, a signatory should also:

adapt city structures, including allocation of sufficient human resources, in order to undertake the necessary actions to take part in developing the Action Plan;

mobilise the civil society.

6

For EU signatories, the recommended baseline year is 1990, or the closest subsequent year for which the most comprehensive and reliable data can be provided. The emission reduction target is set against the baseline year and it can be set either as absolute reduction or per capita reduction. The Baseline Emission Inventory (BEI) covers the CO2 emissions that occur due to energy consumption in the territory of the local authority. The following sectors (often referred to as key sectors of activity) are recommended: - municipal buildings, equipment and facilities; - tertiary (non municipal buildings, equipment and facilities); - residential buildings; - municipal public lighting; - urban road transportation (including municipal fleet, public transport, private transport). The energy-related emissions coming from other sectors might be included in the BEI, if the SEAP foresees measures for them (e.g. industry not under the European Emission Trading Scheme, highways not exploited by the city but on its territory). Some emission sources not related to energy consumption might be also included in the BEI and in the SEAP, for example wastewater and solid waste treatment. The Local Authority may wish to include actions aiming at reducing the CO2 emissions also on the supply side (e.g. development of the district heating network, wind farms, PV, etc…). In this case, local energy (electricity, heat/cold) production should be accounted for in the BEI. The scope of the Sustainable Energy Action Plan (SEAP) is then to define, to describe and to estimate quantitatively energy-related greenhouse gas reduction measures. A large dataset of very different actions proposed by cities has been compiled and can be consulted upon demand. A SEAP should contain both short term actions and mid-long term strategies. The key sectors of activity are the ones, whose inclusion in the BEI is "strongly recommended". Moreover, the local authority is expected to play an exemplary role, by taking outstanding measures on its own buildings, facilities and fleet.

1.3. Proposed adaptation of the methodology for Central Asian Cities

In the specific context of the Eastern Partnership and Central Asian Cities, a new approach is needed in order to allow social and economical progress after the economic collapse in the 1990s. However, the signatories should adopt measures so that this progress occurs in a sustainable manner and that energy efficiency criteria are applied to existing buildings and infrastructures, as well as to all new developments. The first aspect that needs to be considered is the choice of the baseline year: choosing 1990 (as recommended for the EU countries) is not appropriate because of the drastic economic collapse that followed the fall of the Soviet Union, resulting in a CO2 emissions reduction of more than 50% in a few years. Therefore, a recent baseline year will be the recommended choice. On the other side, since the countries covered by this project are recovering from the economic collapse of the 1990s, imposing a reduction of CO2 emissions in absolute terms may not be feasible. Therefore, the opportunity to calculate a target based on a reference scenario, called business-as-usual (BAU) (defined as a continuation of the current trend) to 2020 will be given to signatories willing to do so (besides the absolute and the per capita reduction options). Starting from present data, the BAU scenario will analyse the evolution of energy and emission levels until 2020, under the hypothesis of continuing current trends in population, economy, technology and human behaviour, without the implementation of a SEAP. The target would therefore be calculated compared to the emission levels forecasted by the scenario for 2020. National coefficients that allow estimating the energy consumption in 2020 (starting from real present data) are provided in this report. Signatories will thus develop a simplified BAU scenario, accounting for the attainment of a normal level of quality of the services (streets with public lighting, water supplied every day, etc…). In principle, the key sectors that should be included in the BEI will be the same as in the CoM EU. The same key sectors should be tackled by the set of actions of the SEAP.

7

In this way, signatories will be able to do their emission inventories of the present situation, and estimate which their emissions in 2020 will be. Then they will commit to an emission reduction target based on their projections of emissions for 2020 following the BAU. The factors will be country-specific, calculated both for CO2 and CO2eq (CO2, CH4, N2O using the GWP100metric) in order to allow signatories to choose the approach they prefer. The factors do account for differences between country evolutions and are based on the sectors which are targeted by the CoM (buildings and transport). A global emission projection with EC in-house data and EC – in-house anthropogenic projection model is thereto used. 1.4 Local projections: an alternative?

From open source information, emission projection tools and results are available for local regions, such as e.g. by UN-Habitat and ICLEI. UN-Habitat launched the Sustainable Cities programme, which has some similarities with the CoM initiative. This is also supported by ICLEI (unifying the Local Governments for Sustainability), who is collecting cities scenarios. However, the collection of local scenarios does not provide a global pool of information that is consistent amongst countries. Such collection can not be labelled impartial, uniform and neutral coverage of all countries from the CoM EU and CoM East. For consistency it is needed to use one single tool, in which the scenarios follow projections set by the European Commission, based on a Commissions model and database for emission growth.

8

2. Description of the Business-as-usual (BAU) scenario. The business-as-usual (BAU) scenario indicates that no or just actual measurements are taken into account for the future emission trends and that world energy consumption will be more than doubled in the 2000-2050 period. The emission inventory projections for the coming four decades are calculated, starting from the base year 2005 with the sector-specific growth rates and technology-based emission factors taking into account different abatement measures per regions, in the frame of the FP7 research project CIRCE (www.circeproject.eu), documented by Doering et al (2010). This BAU scenario was also used by the global climate model ECHAM to investigate areas of high air pollution with high health risks for the near future. Pozzer et al. (2012) indicated hot spots with high pollution index per capita for several cities in particular in the Middle East and South East Asia. 2.1 Data and models used As basis the Emission Database for Global Atmospheric Research (EDGAR-CIRCE) was used, which contains global anthropogenic emissions inventories of various air pollutants and greenhouse gases. Based on the EDGARv4version inventories have been calculated for the CIRCE project providing historical (1990-2005) global anthropogenic emissions of the greenhouse gases CO2, CH4 and N2O (the air pollutants and particulate matter are not of importance for the CoM). Emissions trends up to the year 2050 were calculated starting from base year 2005. Activity growth rates from the POLES model (Russ et al., 2007) were used to calculate activity data for the energy sector starting from the EDGAR-CIRCE base year dataset (residential, transport, ships, aviation, transformation and refineries). The POLES (Prospective Outlook for the Long term Energy System) model is also an EC in-house macro-economic model based on partial energy equilibrium. It contains technologically-detailed modules for energy-intensive sectors, including power generation, iron-steel, aluminium and cement production, and transportation sectors to simulate the development of energy scenarios until 2050 on world-scale (for 47 different regions) with one single oil market and three regional gas markets. The growth rates of POLES are differentiated for one power generation sector, three fuel production sectors, three energy consumption sectors, for four transport sectors, for the different fuel types, and for 29 countries and 23 regions. The same growth rates for the fuel consumption were used for the industrial production sector, assuming that the activity/emission trends in industrial production follows the combustion trends in that sector. It should be noted that neither a technology shift nor a fuel shift is explicitly modeled for a given industry sector. The growth rates are entirely based on the economic dynamics of fuel costs and carbon taxes of POLES and the fuel shifts modeled in there. Trends in agriculture, land use and waste are provided from the IMAGE model that is compatible with the POLES baseline and climate stabilization scenarios (P. Russ, D. van Vuuren, personal communication and van Vuuren et al., 2009) with emission trends given by world region for a baseline scenario. The IMAGE model (Integrated Model to Assess Global Environment) is a model of MNP which comprises an Energy-Industry System, a Terrestrial Environment System and an Atmosphere-Ocean System, of which the model one was the most important to project agricultural land-use change, crop production and animal elevation. Also sectors such as ‘use other products’ (including solvents) use the population growth rate from IMAGE (van Vuuren et al, 2009) for the growth rates of the emissions. For the sectors ‘solid waste disposal’ (main sector: waste) emissions of all substances are scaled with the population growth rate, while emissions of ‘waste water treatment’ are scaled with growth rates of sewage. The ‘agriculture’ sectors are treated in different ways. In the sectors ‘agricultural soils’ the emissions of the respective substances are scaled with the specific growth rates of the corresponding emitters, e.g. N2O emissions are scaled with the growth rate of fertilizer combined with the growth rate of crop residues. Emissions of ‘enteric fermentation’ are scaled with the growth rates of the corresponding

9

animals. Emissions of ‘manure management’ of the respective substances are scaled with the growth rates of animal waste. Emissions of ‘agricultural waste burning’ are scaled with the growth rates of CH4 emissions from agricultural waste burning (IMAGE model). Indirect emissions are assumed constant in time. The “Business-as-usual” scenario, hereafter referred to as BAU, explores the situation when no further climate and air pollution policies are implemented beyond what is in place since the year 2005. This means that energy consumption from 2005 to 2050 is driven by population and economic growth but not by energy efficiency/climate change policies (POLES baseline scenario). The combustion technologies/abatement measures are assumed not to change beyond the year 2005 technologies. 2.2 Outcome of the emission projections Table 1 presents global Greenhouse gas (GHG) emissions excluding sectors as biomass burning, international shipping and aviation. In order to compare greenhouse gas emissions of individual gases (here CO2, CH4 and N2O) were converted into CO2-equivalents based on the conversion using the Global Warming Potential (GWP), which express the contribution to global warming of the specific greenhouse gases in relation to carbon dioxide. Throughout this background document the 100 year GWP values as used in the Kyoto Protocol are applied (IPCC, 1995). Applying the Business-as-usual (BAU) scenario global emissions of greenhouse gases increase from 37 Pg CO2-equivalents in 2005 to 66 Pg in 2050 with an important growth (+79 %) due to the continuing increase of the energy sector (in particular for the region East Asia and Southeast Asia) but also to a rather strong contribution of N2O and CH4 emissions (see Table 1). Adaptation and mitigation strategies that extremely influence GHG would be needed to reduce the annually emitted over 35 Pg CO2-equivalents globally with 30 %, which would be needed to guarantee that the temperature will not exceed 2 °C above pre-industrial levels. The latter was investigated with a climate change (CC) scenario. Table 1: Projections under the BAU scenario for global greenhouse gas emissions. (The last row (Italics) indicates the GHG totals in Mt CO2 eq.

Under the BAU scenario strong emission increases occur for both the greenhouse gases (e.g. CO2 +84%) (but also for air pollutants that are similarly as CO2 combustion related such as NOx). The trends for the air pollutants, aerosols and GHG start a fastened growth from 2010 onwards under the BAU scenario. From 2010 to 2025 an increase of 10 % to 30 % is demonstrated and the variation is not only region-specific but also substance- and technology-specific increase because of the expanding sector. Under the BAU scenario the increase for the GHG emissions is about +30 % by 2025, and amounts even to about +70 % by 2050. For the air pollutants CO, BC, NMVOC and SO2 emissions, the increase is about +20 % by 2025 and about +45 % by 2050. For OC and NOx a stronger increase is present of about +60 % by 2050. For the international shipping, all pollutants follow an emission increase for the BAU scenario (2010: +23 %, 2025: +81 %, 2050: +171 %), because mainly of the absence of air pollutant abatement technologies in this sector. A comparison between CIRCE and other references (IPCC, 2005 and 2007 and van Vuuren et al. (2009)) for the CO2, CH4 and N2O emissions in 2000/2005 and projections for 2050 is shown in Fig. 1.

10

The base trend for the period 1990-2005 is described by van Aardenne et al, (2009) and presents an increase of 23 % which is in line with the 24 % increase for the period 1990-2004 reported in the IPCC AR4-WG3-TS (2007). However the emissions of the base year 2005 only represents 75 % of the total emissions considered by IPCC-WG3-TS (2007) because of the absence of the land-use, land-use change and forestry sector (LULUCF) and because of differences in biomass burning. The BAU scenario together with a climate change (CC) scenario are compared to the scenarios SRES A1B and B1p as defined by IPCC in AR3 (2005) under similar conditions as considered for BAU and respective CC in this study. The BAU scenario and the IPCC A1B scenario show for each of the three GHG a similar doubling of the emissions in 2050 relative to 2000, whereas the CC scenario and the IPCC B1p scenario significantly differ. The CC scenario indicates a decrease for all GHG with in total 20 % less emissions in 2050 relative to 2000 but the IPCC B1p scenario further projects a slight increase. In the IPCC AR4 the different scenarios are reviewed taking into account the high uncertainty (>50 %) on the projected GWP and the scenario occurrence probability, which both lead to a subdivision into VI categories. The relative growths of GHG emissions in the BAU and CC projections are confronted with scenarios results (SRES I to VI) of the IPCC-AR4-WG3-TS (Table TS.2) (2007). As it is shown in Fig. 1 emissions projections of SRES category V and A1B are in the same range as the BAU CIRCE projections. The same trend of the CC scenario is apparently observed in the emissions projections of SRES category II.

Fig. 1: Total GHG emissions (2000 and 2005/2004) and projections for 2050 under global Business-as-usual (BAU) and Climate Change (CC): CIRCE scenarios compared with other references (IPCC – 2005, IPCC - 2007, van Vuuren et al. (2009).

11

3. National coefficients derived from a business-as-usual projection for the CoM East countries. 3.1 Results for CO2 and CO2 eq national projections Based on the study described in section 2 the BAU factors of Table 2 were derived for the CoM-East countries. This results from using the projections done with the POLES model, and selecting the sectors which are relevant for the CoM (buildings and transport). For each country and year of baseline emission inventory, there is the factor for CO2 according to the BAU projections to 2020. Depending of the country and year selected, these factors will lead to emissions in 2020 up to 26% higher than the emissions in 2005. However, for many countries, these factors are < 20%, and as such, still requiring a decrease in emissions under the 20% reduction target. This will be reasonable for some countries which already have high per capita emissions (Ukraine, Kazakhstan, Turkmenistan) but not that sensible for the ones with important agricultural activities and lower per capita emissions (Georgia, Kyrgyzstan, Moldova, Tajikistan). In the following table you can see the per capita emissions for all these countries (1990, 2005, 2005 and 2008) from the EDGAR database, version 4.2 http://edgar.jrc.ec.europa.eu (this includes all sectors, not only Covenant sectors). There is also an estimation of the emissions in 2020 based on those of 2008 and using the factors we have calculated. The per capita emissions do not change fast over a decade. For those countries with low per capita emissions (<3 ton CO2/cap), the per capita emissions in 2020 will remain low (<3 ton CO2/cap), while Kazakhstan will be allowed to increase their per capita emissions up to 15.6 ton CO2/cap, which is almost the double of the mean EU-27 per capita emission. A further refinement of the national coefficients deemed necessary, looking at the local effects and as such differentiating between rural and urban areas. This is done in the following section 4. Table 2: Overview of the CO2 per capita for the 11 post-Soviet countries in 1990-2000-2005 and 2008 and the BAU factors which should present a continuation of the trend. These CO2 factors are focused 2020 and the Covenant of Mayors sectors only.

Similar to CO2 also projections were done for CH4 and N2O, which yielded CO2eq factors. Figures 2 & 3 underneath represent the Business-as-usual factors which CoM-East signatories could use to estimate their emissions in 2020 based on their current (or recent emissions). These factors have been calculated using the emissions projections from the POLES model for the buildings and transport sectors. The factor for each year and country is calculated from the ratio between projected emissions in 2020 and emissions in each of the other years. These factors are given for inventories of CO2 only (Fig.1) or of inventories of the greenhouse gas (GHG) with CO2-eq.

12

Fig. 2: National coefficients for CoM-East signatories to estimate their CO2 emissions in 2020 based on their current (2005-2020) emissions estimates for the buildings & transport sectors. These coefficients are derived as the ratio of the projected CO2 in 2020 to the CO2 in a preceding baseyear.

Fig.3: National coefficients for CoM-East signatories to estimate their GHG emissions in 2020 based on their current (2005-2020) emissions estimates for the buildings & transport sectors. These coefficients are derived as the ratio of the projected CO2-equivalent (considering CO2, CH4 and N2O emissions in the IPCC metric of GWP100) in 2020 to the CO2-equivalent in a preceding baseyear. 3.2 Discussion of the limitations of the projections It should be noted that the projections are done with global growth rates, taking into account the historical trends in the NIS countries from 1990-2005. The projection is a global one (because of the coupling of fuel markets globally), based on IEA data (2007), following an EC internal scenario

13

(developed by DGENV and JRC-IPTS for the climate-energy package of Europe 2020). The advantage is that the projections can be defended for all countries equally with one single methodology, consistently applied. Therefore the results are not only valid for the 11 countries of the Eastern Partnership but also for EU-27 itself and for Europe's Southern Partnership etc). Moreover the projections are done for all sectors, energy-related and agricultural related sectors (the latter is in particular important when including non-CO2 gases such as CH4 and N2O. Not only all greenhouse gases, but also all other air pollutants and aerosols were considered, this allowing a global assessment of climate and air quality changes on citizen’s health and wealth. The large difference in emission per capita between some countries (e.g. Georgia, only 1.2 tonCO2 /capita and Kazakhstan 15 tonCO2 /capita) is inherent to the traditional choice of fuel. In all Newly Independent States the residential sector is burning largely primary solid biomass (wood, wood waste, vegetable waste, ...), which is not contributing to the annual CO2 emission total. (The CO2 the plants took up during the growing year, is released at the combustion, resulting in a zero emission release.) In Georgia, a fuel share smaller than 25% is gas and even a smaller one is oil. In Kazakhstan a larger fuel share of about 40% is patent fuel. That explains directly that Kazakhstan in the residential sector is emitting much more and invite for a potential refinement of the allowed BAU growth rate projections with extra criteria based on that CO2 per capita. Comparing the national factors from 2005-2010 (result of the first five years projection) with the current IEA energy consumption data for the same period, we recognise significant discrepancies for Georgia. There are several reasons for that: (i) The IEA data are not varying smoothly (e.g. in natural gas for Georgia in the residential sector one sees that the natural gas consumption from 1999 to 2000 doubled and then halved again by 2002); (ii) Recessions cannot be predicted and are known to have an influence, sometimes with some time-lag between countries; (iii) The latest IEA data statistics are subject to revision with corrections backwards in time; (iv) As shown by Paruolo et al (2012), an econometric analysis of the full historical emission trends can neither confirm a direct relation between the levels of emission and levels of income, nor the causality. Finally a global increase of CO2 with about 10% over 15 years (from 2005-2020) is already a respectable increase. This order of increase is also found back from 1990-2005 (12%) as reported by Olivier et al (2011).

14

4. The urban dimension The UNDP provides statistics on their website http://hdrstats.undp.org/en/indicators/default.html addressing human development. One of the human development indicators listed by UNDP, is the CO2 emissions per capita, provided by Boden et al (2009). Their database compares for the CO2 very well with the EDGAR database. The final composite Human Development Index, HDI, is composed of improvements in health (life expectancy at birth), education (years of schooling) and living standards (with income per capita). The HDI facilitates instructive comparisons of the experiences within and between different countries. It is observed that the development starts in the cities, where more human activities are concentrated. The development in the countryside lags behind the development in the city, which is one of the reasons for the urbanisation process that is ongoing in many countries. In this section we diversify the Business-as-usual growth rate of a country for urban and rural regions. For regions where the economic growth took off considerably, we assume that growth rates apply for the global population, but for countries which are still in the initial phase of economic growth, it is assumed that first the economic growth will take place in the urban areas and that the rural areas are not yet playing a relevant role. The 11 countries of EC’s Eastern Partnership have been screened with regard to their economic development and Table 3 lists whether UNDP considers these countries still in the initial phase of economic growth or in a more developed phase.  Table 3:  Human development index and GHG emission level per capita for the 11 post‐Soviet countries for the 

period 2005‐2011 according to UNDP data and EDGARv4.2 

Since the baseyear for the projection is 2005, and Belarus, Georgia, Kazakhstan and Ukraine already were considered to have a high human development index, it is considered less appropriate to apply an urban factor on these 4 countries (BLR, GEO, KAZ and UKR) in a first step. Nevertheless, the evolution from 2005 to 2010 changed considerably, so that the selection of countries based on HDI is quite sensitive to the year of reference. Therefore a more pragmatic consideration of the emission level was opted for and we recommend to leave out BLR, KAZ, UKR and also TKM for the application of the urban factor, but not GEO.

15

4.1 . Distinction rural versus urban population To distinguish between urban and rural population, information on the population density per gridcell of 0.5km by 0.5 km was used. For a country the changes in population density (from one grid cell to another) was analysed, assuming an S-shaped curve between lowest populated and most dense populated regions. Country-specific thresholds for urban areas are calculated consistently for all world countries applying the same algorithm with optimisation process. EDGAR uses population data from CIESIN, http://sedac.ciesin.columbia.edu/gpw/global.jsp , to derive proxy data representing globally population density, with a diversification of rural and urban population density. The CIESIN site provides Population grids from 1990 to 2015 (5 years step) at different resolutions. For our purposes the resolution 2.5’ (minutes) has been chosen. The re-gridding to 0.1o (degrees) has been performed using ArcMap and ad-hoc PHP programs. In addition, CIESIN provides a Settlements Points grid at a resolution of 30’’ (seconds). This grid is the reference for the computation of urban/rural population coefficients. The method to distinguish between rural and urban population is based on the assumption that the quota of population to be considered as Urban. The quota of Urban population depends on the population density in grid cells following a linear function for densities between specific ranges until a defined saturation point after which the population of the cell is considered completely Urban, while before the low value of the range the population is considered completely rural. This is visualised in Figure 4.

Fig. 4a: The low_value (LV) and high_value (HV) are evaluated country by country using the Urban Extent map for the year 2000 provides by CIESIN. Those HV and LV values give indication of the level of adaptability of the population of a country, i.e. when a cell is assigned an Urban qualifier with a low HV density (say 800 to 1200) then it is normally an industrialized country that provides urban infrastructures even for a scarce population.

The Urban Extent map provided by CIESIN is a set of small cells (30’’) indicating the presence or not of settlements in the cell. Using ArcGis this map has been Resampled to 0.01o , to be a multiple of our population maps, then mapping 100 settlement cells per each population cell (0.1o) the total number of ‘settled’ cells gives the percentage of urbanization of the 0.1o population cell. After these operations we have a single urbanization map for the year 2000 that leads to an Urban Population and a Rural Population map for the year 2000.

The method consists in deriving the two function coefficients (LV and HV) for each country by computing the rural population for the year 2000 iterating different combinations of coefficients, until the couple of

16

coefficients leads to the minimum Standard Deviation of the difference between the computed rural population and the rural population obtained from the Urban Extent map. The iteration process is visualised in Fig. 4b. The iteration values for LV are 10 to 600 stepping by 10, while for HV are 500 to 1900 stepping by 50 units of heads/km2. In order to avoid divergence in the minimization process, the points exceeding 2*sigma are eliminated from the process.

Fig. 4b : Iterative process combinations of coefficients, to find the (LV, HV) couple of coefficients that

leads to the minimum Standard Deviation of the difference between the computed rural population and the rural population obtained from the Urban Extent map of CIESIN.

Figure 5 shows the trends of the Standard Deviation of the difference for a country computed for three low_values (LV: 10, 200, 390) and different High_values (HV). The minimum sigma occurs for LV=390 and HV=1300 heads/Km2 in the case of a country under development (Fig.5.a: China) and at LV=10 in the case of a developed country (Fig. 5.b: USA).

Fig. 5a : Trends of the Standard Deviation of the difference for China computed for three low_values (LV: 10, 200, 390) to derive minimum sigma. Fig. 5b: Trends of the Standard Deviation of the difference for USA computed for three low_values (LV: 10, 390, 180) to derive minimum sigma.

As a result Figure 6 shows the rural and urban population for northern India and Pakistan. Blank spots are cells where rural population is zero, i.e. cities. The urbanization process can be noticed in northern India and Pakistan.

17

Rural Population in northern India in 1970 Rural Population in northern India in 2015

Fig. 6: Rural population in northern India and Pakistan in evolution from 1970 to 2015 with a decreasing trend because of the urbanisation process.

4.2. The resulting urban factor for the Eastern Partnership countries Using the data included in the EDGAR database, in relation to urban and rural population, in 2005, 2010 and 2015 for these countries, urban factors Xcity are derived under the assumption that the economy will further emerge first in the cities and that most of the increase on emission production from 2005 to 2020 will take place in urban places rather than in rural communities. Therefore, we propose to use the following formula to estimate the emission increase which will take place in cities: Xcity = (Xcountry * Pop_tot2005 -Pop_rur2005)/Pop_urb2005 where Xcountry is the factor calculated for countries for a certain year, Xcity is the factor adapted for a city in a certain country and Pop_tot, Pop_rur and Pop_urb are total, rural and urban populations for a certain country in a certain year. This formula allocates for each country its business-as-usual growth factor to its urban population, expressed by the criteria that the national factor multiplied with the total population equals the new city factor multiplied with the urban population plus the factor 1 (steady state) multiplied for the rural population. As populations are only available until 2015, these factors have only been calculated for these 3 years, and taking into account the projections of urban to rural ratios until 2015, as presented in Table 4. Table 4: Overview of the urban factor for each of the 11 post-Soviet countries, based on the origin-nal national factor and the urban/rural population in 2005-2010-2015.

18

Those countries having originally high per capita emissions (BLR, KAZ, TKM, UKR) already started with the emerging of their economy and increase in emissions longer time ago. Therefore the assumption that further economic growth is driven by growth in the city is not realistic. Therefore we do not recommend for these countries to use urban factors, but we recommend the national factors instead. (In the case of TKM and UKR, the original factors and the urban factors are very similar and show not greater increases.) For all other 7 countries with the emerging of their economy in a much earlier phase the expectation that the development is greater way in cities rather than in rural areas can be considered realistic. These ratios from urban to rural population allow us to provide city factors for any year signatories may want to choose as baseline year from 2005 to 2020. They are derived with the annual country factors previously calculated for CO2 and CO2eq, and the urban and rural populations for 2005 for the period 2005-2009, population data for 2010 for the period 2010-2014 and population data for 2015 for the period 2015-2020. The obtained city factors can be seen in the following Table 5 for CO2 and Table 6 for all greenhouse gases. Table 5: Urban coefficients for CoM-East signatories projecting their CO2 emissions in 2020 based on their current (2005-2020) emissions estimates for the buildings & transport sectors. These coefficients are derived as the ratio of the projected CO2 in 2020 with the urban population trends 2005-2015 to the CO2 in a preceding baseyear.

Table 6: Urban coefficients for CoM-East signatories projecting their greenhouse gase emissions in 2020 based on their current (2005-2020) GHG emissions estimates for the buildings & transport sectors. These coefficients are derived as the ratio of the projected GHG emissions in CO2equivalent in 2020 with the urban population trends 2005-2015 to the CO2equivalent in a preceding baseyear.

19

It is recognised that almost identical factors apply for the CO2 and for the greenhouse gases in CO2 equivalent. For the sake of completeness, the national coefficients for CO2 and CO2-eq from Fig 2 and 3 under section 3 are updated to the city projections of Fig. 7 and 8.

Fig. 7: Country-specific urban coefficients for CoM-East signatories to estimate their CO2 emissions in 2020 based on their current (2005-2020) emissions estimates for the buildings & transport sectors. These coefficients are derived as the ratio of the projected CO2 in 2020 and the urban population to the CO2 in a preceding baseyear under a business-as-usual scenario.

Fig. 8: Country-specific urban coefficients for CoM-East signatories to estimate their GHG emissions in 2020 based on their current (2005-2020) emissions estimates for the buildings & transport sectors. These coefficients are derived as the ratio of the projected CO2eq in 2020 and the urban population to the CO2eq in a preceding baseyear under a business-as-usual scenario.

20

4.3. Discussion and limitations of the use of the urban factor The aim of this section is to discuss the use of the urban factor. We first present a comparison of how these business-as-usual projections done on a country basis would look if they were adapted to a certain city (ie. Tbilisi in Georgia) with the aim of comparing the obtained results/factors with those used by the city of Tbilisi to do their SEAP. In a second step a recommendation for using the urban factor for any city from the selected countries of the Eastern Partnership is given. The Business-as-Usual (BAU) scenario for Tbilisi (capital of Georgia) by NATELI (2011) and its use for the Sustainable Energy Action Plan prepared by Tbilisi City (2011) used a local prospective outlook model for energy scenarios LEAP (Long range Energy Alternatives Planning System). Even though this tool, focusing on Georgia and its capital might give very reliable prediction for Tbilisi, its results can neither be generalised for all cities of the Eastern Partnership countries nor assumed available for any other city without extra funding (such as Tbilisi was receiving from USAID). Its results nevertheless compare within 6% difference with the BAU-scenario coefficient of JRC-IES for Georgia, corrected with the urban factor. Contrary to the LEAP model, the JRC-IES scenario is based on a Commission prospective outlook tool for energy scenarios (POLES) and valid for all world countries. We conclude this chapter with some more general recommendations:

Accurate, complete and up-to-date baseyear emissions are primordial: the more complete and recent, the less biases are present at the start of the projections. The inventory making of emissions takes time and a delay of one to two years is common. A baseyear in the period between 2005 and 2010 is therefore considered feasible and recommended.

A first indication on how much the country contributes to the global warming is obtained by

comparing the emission level per capita (cfr. Annex III) with the European mean average of 9.9CO2eq/cap in the case of all GHG and 8.1CO2/cap in the case of only CO2. The emissions levels per capita for Belarus, Kazakhstan, Turkmenistan and Ukraine exceed the mean EU-27 average level. Therefore extra growth for emissions by BAU coefficients, and in particular BAU coefficients multiplied with an urban factor, have to be questioned.

The country-specific coefficients for CoM-East cities are most coherently derived when using a global prospective outlook model addressing all countries. As such no fast and local variations, such as recessions are perturbing the projections for 2020.

A distinction between the countries based on their economic growth seems necessary. For those countries which are no longer in the initial phase of economic development (and have higher HDI and high GHG emission levels), the urban factors are not applicable.

All cities of one single country are allowed to use the same BAU coefficient, eventually

corrected with the urban factor, as set for the country, in order to obtain a national consistency.

This exercise can be repeated for all world countries, to give a more equilibrated view on the cities which are already signatory in the CoM of Western-European countries and on cities which could join a similar CoM designed for countries in other EC Partnerships, such as CoM South.

21

5. Conclusion On the basis of the work carried out for the CIRCE project we have obtained projection factors to be used to estimate greenhouse gas emission in 2020 for 11 countries for the Eastern Partnership for the Covenant of Mayors under a business-as-usual scenario. An additional assessment has been carried out to estimate factors which could be adapted for cities, assuming that most of the emission increase in those countries will take place in cities. It was observed that these factors are very country-specific, but are not very much dependent on the substance. Moreover it was considered that the emission growth is realistically mainly present in cities for these Eastern Partnership countries, except for Belarus, Kazakhstan, Turkmenistan and Ukraine. For the latter 4 countries a national coefficient is recommended instead. Figure 6 provides an overview of the recommended country-specific coefficients for CoM-East signatories to estimate their CO2 or CO2eq emissions in 2020 based on their current (2005-2020) CO2 or CO2eq emissions estimates for the buildings & transport sectors. The values are also summarised in table 7.

Fig. 6: Country-specific coefficients for CoM-East signatory cities to estimate their CO2 or GHG emissions in 2020 based on baseyear (2005-2020) estimates for the buildings & transport sectors. Table 7: Summary of the country-specific coefficients for CoM-East signatories to estimate their CO2 or GHG emissions in 2020 based on baseyear estimates for the buildings & transport sectors.

22

References Doering, U.M., U.M. Doering, G. Janssens-Maenhout, J.A. van Aardenne, V. Pagliari (2010), CIRCE report

D.3.3.1, Climate Change and Impact Research in the Mediterranean Environment: Scenarios of Future Climate Change IES report 62957.

Pozzer, A., P. Zimmermann, U.M. Doering, J. van Aardenne, H. Tost, F. Dentener, G. Janssens-

Maenhout, and J. Lelieveld, Effects of business-as-usual anthropogenic emissions on air quality, Atmos. Chem. Phys. Discuss., 12, 8617-8676, 2012, doi:10.5194/acpd-12-8617-2012

Schade, B., Wiesenthal, T. (2007), Comparison of long-term world energy studies. Assumptions and results from

four world energy models, IPTS EUR report 22398 EN. Shindell, D., Kuylenstierna, J.C., Vignati, E., van Dingenen, R., Amann, M., Klimont, Z., Anenberg, S.C.,

Muller, N., Janssens-Maenhout, G., Raes, F., Schwartz, J., Faluvegi, G., Pozzoli, L., Kupiainen, K., Höglund-Isaksson, L., Emberson, L., Streets, D., Ramanathan, V., Hicks, K., Oanh, N.T.K., Milly, G., Williams, M., Demkine, V., Fowler, D. (2012) Simultaneously mitigating Near-Term Climate Change and Improving Human Health and Food Security, Science, Vol. 335, p. 183-189

United Nations Dep. of Econ. & Soc. Affairs/ Population Division (UNDP) (2005), World Population

Prospects: The 2004 Revision, No. E.05.XIII.12

IPCC, 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Eggleston, S., Buendia, L., Miwa, K., Ngara, T., Tanabe, K. (Eds). IGES, Japan.

OECD/IEA (2010), CO2 Emissions from fuel combustión, 2010 Edition, Part III.1-49: Olivier, J.G.J., Janssens-Maenhout, G., van Aardenne, J., Greenhouse gas emissions: Shares and trends

Olivier, J.G.J., Van Aardenne, J.A., Dentener, F., Ganzeveld, L. and J.A.H.W. Peters 2005. Recent trends in global greenhouse gas emissions: regional trends and spatial distribution of key sources. In: Non-CO2 Greenhouse Gases (NCGG-4), A. van Amstel (coord.), page 325-330. Millpress, Rotterdam, ISBN 905966 043 9.

Olivier, J.G.J., Janssens-Maenhout, G., Peters, J.A.H.W., Wilson, J., Long-Term Trend in Global CO2 Emissions: 2011 Report – JRC-PBL report, ISBN 978-90-78645-68-9, http://edgar.jrc.ec.europa.eu/news_docs/C02%20Mondiaal_%20webdef_19sept.pdf

Paruolo, P., Murphy, B., Janssens-Maenhout, G. (2012) “Do emissions and income have a common trend? A global time-series analysis 1970-2008, WP Facoltà di Economia 2011/12, Università dell' Insu- bria Varese, Italy, available at http://eco.uninsubria.it/dipeco/quaderni/files/QF2011_13.pdf Russ, P., Wiesenthal, T., van Regenmorter, D., Ciscar, J. C., 2007. Global Climate Policy Scenarios for 2030

and beyond. Analysis of Greenhouse Gas Emission Reduction Pathway Scenarios with the POLES and GEM-E3 models, JRC Reference report EUR 23032 EN (http://ipts.jrc.ec.europa.eu/publications/pub.cfm?id=1510)

Van Vuuren, D.P., den Elzen, M.G.J., van Vliet, J., Kram, T., Lucas, P., Isaac M. (2009), Comparison of different climate regimes : the impact of broadening participation, Energy Policy, Vol. 37, p. 5351-5362

United Nations Dep. of Econ. & Soc. Affairs/ Population Division (UNDP) (2011) Human Development Report 2011- Sustainability and Equity: A Better Future for All, http://hdr.undp.org/en/media/HDR_2011_EN_Summary.pdf Boden, T.A., G. Marland, and R.J. Andres. 2009. Global, Regional, and National Fossil-Fuel CO2

Emissions. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tenn., U.S.A. doi 10.3334/CDIAC/00001

23

NATELI (2011), New Applied Technology, Efficiency, Lighting Initiative Tbilisi: Model and Business as Usual (BAU) Scenario for Tbilisi City Sustainable Energy Action Plan (SEAP), prepared for USAID/Causcasus, June 2011

Tbilisi City (2011), Sustainable Energy Action Plan for 2011-2020, approved by the government of

Tbilisi City on 28 March 2011, Decision No. 07.10.237

24

Annex I: Ranking of world countries by GHGs EDGARv4.2 provides its results of Greenhouse Gas (GHG) emission totals for all world countries as below. Please consult also http://edgar.jrc.ec.europa.eu/overview.php?v=intro&sort=des8.

25

26

27

28

Annex II: Ranking of world countries by CO2/cap EDGARv4.2 provides its results of CO2 emission totals per capita for all world countries as below. Please consult also online the list on http://edgar.jrc.ec.europa.eu/overview.php?v=CO2&sort=des8

29

30

31

32

33

Annex III: Ranking of world countries by CO2eq/cap EDGARv4.2 provides its results of GHG emission totals per capita for all world countries as below. Please consult also list on http://edgar.jrc.ec.europa.eu/overview.php?v=CO2eq&sort=des8

34

35

36

37

38

List of Abbreviations AZE Azerbaijan

ARM Armenia

BLR Belarus

GEO Georgia

KAZ Kazakhstan

KGZ Kyrgyzstan

MDA Moldova

TJK Tajikistan

TKM Turkmenistan

UKR Ukraine

UZB Uzbekistan

ADAM ADaptation And Mitigation Strategies project

BAU Business-As-Usual

BEI Baseline emission inventory

BC Black carbon

CC Climate Change

CIRCE Climate Change and Impact Research in the Mediterranean Environment

CO2eq CO2 Equivalent

CoM Covenant of Mayors

CoM East Covenant of Mayors Eastern Partnership

DGENV Directorate General for the Environment

EDGAR database Emission Database for Global Atmospheric Research

EC European Commission

GWP Global Warming Potential

GWP100 Global Warming Potential calculated for a period of 100 years

GHG Greenhouse Gas

IEA International Energy Agency

ICLEI International Council for Local Environmental Initiatives

IMAGE Integrated Model to Assess Global Environment

IPCC Intergovernmental Panel on Climate Change

IPCC A1B IPCC Scenario type A1B

IPCC B1p IPCC Scenario type B1p

IPCC AR3 IPCC Third Assessment Report

IPCC AR4-WG3-TS IPCC Fourth Assessment - Technical Support Unit of Working Group III

IPCC-WG3-TS IPCC - Technical Support Unit of Working Group III

39

JRC-IPTS Joint Research Centre - Institute for Prospective Technological Studies

LULUCF Land-Use, Land-Use Change and Forestry Sector

MNP Netherlands Environmental Assessment Agency

NIS countries Newly Independent States

NMVOC Non-methane volatile organic compounds

OC Organic carbon

POLES Prospective Outlook for the Long Term Energy System

SRES Special Report on Emissions Scenarios

SEAP Sustainable Energy Action Plan

 

2000-AR4 Projections according to IPCC SRES Fourth Assessment Scenario for 2000

2000-CIRCE Projections according to CIRCE Scenario for 2000

2004- AR4 Projections according to IPCC SRES Fourth Assessment Scenario for 2004

2005-CIRCE Projections according to CIRCE Scenario for 2005

2050-ADAM Projections according to ADAM project for 2050

2050-AR3-A1B Projections according to IPCC SRES Third Assessment Scenario Type A1B for the year 2050

2050-AR3-B1p Projections according to IPCC SRES Third Assessment Scenario Type B1p for 2050

2050-AR4-catII Projections according to IPCC SRES Fourth Assessment Scenario cat II for 2050

2050-AR4-catV Projections according to IPCC SRES Fourth Assessment Scenario for 2050

2050-CIRCE-BAU Projections according to CIRCE Business as Usual Scenario for 2050

2050-CIRCE-CC Projections according to CIRCE – Climate Change Scenario for 2050

40

41

How to obtain EU publications Our priced publications are available from EU Bookshop (http://bookshop.europa.eu), where you can place an order with the sales agent of your choice. The Publications Office has a worldwide network of sales agents. You can obtain their contact details by sending a fax to (352) 29 29-42758.

42

The mission of the JRC is to provide customer-driven scientific and technical support for the conception, development, implementation and monitoring of EU policies. As a service of the European Commission, the JRC functions as a reference centre of science and technology for the Union. Close to the policy-making process, it serves the common interest of the Member States, while being independent of special interests, whether private or national.

LB

-NA

-25315-EN

-N

ISSN 1831-9424 ISBN 978-92-79-24803-0 doi:10.2788/26047